Grok Imagine - AI Image & Video Generator

Grok Imagine is an xAI video model on PicX Studio for generating video from text, images, and reference media.

Yes. Add an image in the PicX Studio generator, then describe the motion, camera direction, and atmosphere you want to create.

Grok Imagine supports video durations from 4 to 15 seconds. Reference clips go up to 10 seconds, and extend adds 2 to 10 seconds to an existing clip.

Yes. Grok Imagine supports native audio generation as part of the video workflow. Every clip comes back with sound.

Yes. Use the extend workflow to keep developing an existing Grok Imagine clip by 2 to 10 seconds at a time.

Your first image is free after you sign up, no card needed. Video models, including Grok Imagine, need a plan or a one-time credit pack. Credits never expire.

On PicX Studio, Grok Imagine is the video model. Make a still with Nano Banana Pro or GPT Image 2, then animate it with Grok Imagine.

Up to seven in one generation. Reference clips can be up to 10 seconds long.

32 credits per second at 480p and 53 credits per second at 720p, with audio included. A 10 second 720p clip is 530 credits.

Grok Imagine is xAI's video model on PicX Studio. It generates videos from text, images and reference images, always with native audio, and it can extend an existing clip. Clips run from 4 to 15 seconds at 480p or 720p in seven aspect ratios. xAI also offers image generation under the Grok Imagine name; on PicX Studio, Grok Imagine is the video model, and you can make the starting image with any of the image models first.

Choose Grok Imagine in the studio and write a concise prompt describing the subject, action, setting, style and sound. Optionally add an image to animate, or up to seven reference images to build a scene from. Pick a length, an aspect ratio and 480p or 720p, then generate. Compare the results, download the one you want, or extend it to keep the scene going.

Image to video keeps the look of your picture and adds motion and sound. Upload one image, then describe what should move, how the camera travels and what you should hear. Leave the aspect ratio on auto to keep the image's own shape. A clear subject, simple background and good light give the steadiest result. If you do not have an image yet, make one with Nano Banana Pro or GPT Image 2 and bring it over.

Reference mode builds a new scene from several images at once: a character from the front and side, an outfit, a product, a location. Grok Imagine uses them as guidance rather than as the first frame, so the camera and the action can be new while the people and things stay recognisable. Reference clips run up to 10 seconds. Name each reference in the prompt, for example "the woman in image 1 holds the bottle from image 2".

Extend continues a video from its last frame. Pick the clip, describe what happens next and choose how much to add, from 2 to 10 seconds. Resolution and framing follow the source clip. Extending in steps gives you control over each beat of a scene, and lets you stop as soon as the story lands instead of paying for one long take that drifts.

Write the scene as one shot: who is in it, what they do, where it happens and how the camera moves. Grok Imagine always generates audio, so add the sound too, such as wind, footsteps or a short line of dialogue in quotes. Keep to one action per clip and use extend for the next one. Describe what you want to see rather than listing what to avoid. Browse the templates for tested Grok Imagine prompts you can copy.